you could have APR::SubProcess take the same args as the apr C
functions. and have an Apache::SubProcess wrapper that takes $r and
passes the required values to the APR::SubProcess function(s).
i have no idea if APR::SubProcess would ever be useful outside of
httpd. but its better to expect the
>>Otherwise how do you get subprocess_env?
>>
>
> my $table = $r->subprocess_env;
[snip]
> if that doesn't help or make sense, give me more context. what is the C
> prototype of the function that needs r->subprocess_env?
That's not what I meant. In APR::SubProcess::spawn_proc_prog I need t
On Fri, 14 Dec 2001, Stas Bekman wrote:
> I see what you mean. A module APR::Foo can be used outside of Apache if
> it doesn't use any ap_ functins. But for example $r->subprocess_env
> needs $r, which is Apache::Request. Doesn't it tie the module to Apache?
huh? no.
> Otherwise how do you
Doug MacEachern wrote:
> On Fri, 14 Dec 2001, Stas Bekman wrote:
>
>
>>ok, that will work for APR::PerlIO, but APR::SubProcess needs $r/$s for
>>doing things like subprocess_env.
>>
>
> $r->subprocess_env is an APR::Table, so no problem there.
> why does it need $s ?
>
> if you're saying that
On Fri, 14 Dec 2001, Stas Bekman wrote:
> ok, that will work for APR::PerlIO, but APR::SubProcess needs $r/$s for
> doing things like subprocess_env.
$r->subprocess_env is an APR::Table, so no problem there.
why does it need $s ?
if you're saying that something really needs a request_rec or ser
Stas Bekman wrote:
> Doug MacEachern wrote:
>
> > On Fri, 14 Dec 2001, Stas Bekman wrote:
> >
> >
> >>Is there a way for SourceScan to tell where it borks? I mean on which
> >>file (line?).
> >>
> >
> > not that i know of, that'd be up to C::Scan.
>
>
> more problems with C::Scan:
>
>
Doug MacEachern wrote:
> On Fri, 14 Dec 2001, Stas Bekman wrote:
>
>
>>I've moved it to APR, but now I'm not sure if it's correct because both
>>APR::PerlIO and Apache::Request need $r or $s, which makes them
oops, s/Apache::Request/APR::SubProcess/
>>dependent on mod_perl. Or is it cor
Doug MacEachern wrote:
> On Fri, 14 Dec 2001, Stas Bekman wrote:
>
>
>>Is there a way for SourceScan to tell where it borks? I mean on which
>>file (line?).
>>
>
> not that i know of, that'd be up to C::Scan.
more problems with C::Scan:
#ifndef MP_SOURCE_SCAN
#include "apr_optional.h"
On Fri, 14 Dec 2001, Stas Bekman wrote:
> I've moved it to APR, but now I'm not sure if it's correct because both
> APR::PerlIO and Apache::Request need $r or $s, which makes them
> dependent on mod_perl. Or is it correct to say that $r/$s doesn't make
> the code dependent on mod_perl?
just
>>But since Apache::SubProcess could be used outside of mod_perl, this is
>>probably not good enough.
>>
>
> if that's the case, it should be APR::SubProcess. it should only be named
> Apache::SubProcess if it references ap_ symbols.
I've moved it to APR, but now I'm not sure if it's correct
Doug MacEachern wrote:
> you can put #ifndef MP_SOURCE_SCAN around the trouble makers.
that works. Thanks.
Is there a way for SourceScan to tell where it borks? I mean on which
file (line?).
_
Stas Bekman JAm_pH
It also explodes on:
#include "apr_optional.h"
_
Stas Bekman JAm_pH -- Just Another mod_perl Hacker
http://stason.org/ mod_perl Guide http://perl.apache.org/guide
mailto:[EMAIL PROTECTED] http://ticket
I've these function declarations in apr_perlio.h:
PerlIO *apr_perlio_apr_file_to_PerlIO(pTHX_ apr_file_t *file,
apr_pool_t *pool);
APR_DECLARE_OPTIONAL_FN(PerlIO *,
apr_perlio_apr_file_to_PerlIO,
(pTHX_ apr_
Apache-Test assumes that any core file found under t was generated by
httpd, (end suggests the gdb invoking command) which is wrong. I
should've read the notice printed by gdb:
"Core was generated by `/home/stas/perl/ithread/bin/perl
/home/stas/apache.org/mp-subproc/t/TEST -confi"
This one was
On Fri, 14 Dec 2001, Stas Bekman wrote:
just use this instead: 'use 5.8.0;'
which will run before ::load is called.
> Also what was the trick to avoid the bootstrap if tried outside of
> modperl env?
the generated code spits out Apache::XSLoader::load for Apache classes and
APR::XSLoader::load
Doug MacEachern wrote:
> On Fri, 14 Dec 2001, Stas Bekman wrote:
>
>
>>I'm not satisfied with autogenerated
>>WrapXS/Apache/SubProcess/SubProcess.pm, since I need to require
>>APR::PerlIO in it, and I don't really want to create the SubProcess.xs
>>and Makefile.PL in xs/Apache/SubProcess/. D
On Fri, 14 Dec 2001, Stas Bekman wrote:
> I'm not satisfied with autogenerated
> WrapXS/Apache/SubProcess/SubProcess.pm, since I need to require
> APR::PerlIO in it, and I don't really want to create the SubProcess.xs
> and Makefile.PL in xs/Apache/SubProcess/. Do you think it's OK for
> Wrap
On Fri, 14 Dec 2001, Stas Bekman wrote:
> - make WrapXS more portable
+1
-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
I'm not satisfied with autogenerated
WrapXS/Apache/SubProcess/SubProcess.pm, since I need to require
APR::PerlIO in it, and I don't really want to create the SubProcess.xs
and Makefile.PL in xs/Apache/SubProcess/. Do you think it's OK for
WrapXS to check whether foo.pm already exists and not t
- make WrapXS more portable
Index: lib/ModPerl/WrapXS.pm
===
RCS file: /home/cvs/modperl-2.0/lib/ModPerl/WrapXS.pm,v
retrieving revision 1.37
diff -u -r1.37 WrapXS.pm
--- lib/ModPerl/WrapXS.pm 2001/12/14 04:10:20 1.37
+++ l
On Thu, 13 Dec 2001, Stas Bekman wrote:
> I need Apache->gensym (actually I need the C function modperl_gensym), I
> wanted to port it from 1.x but todo says:
>
> Apache->gensym:
> >> Apache::compat ?
>
> we don't need it anymore? I know we don't need it for the Perl side, but
> we do need i
On Fri, 14 Dec 2001, Stas Bekman wrote:
> - don't cache the autogeneration warning (it's not the same because of
> the generation trace)
+1
-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL
- don't cache the autogeneration warning (it's not the same because of
the generation trace)
Index: lib/ModPerl/WrapXS.pm
===
RCS file: /home/cvs/modperl-2.0/lib/ModPerl/WrapXS.pm,v
retrieving revision 1.36
diff -u -r1.36 WrapXS.pm
23 matches
Mail list logo
